QBiC

Project Members:

Prof. Dr. Verena Keitel

verena.keitel@med.uni-duesseldorf.de

Universitätsklinikum Düsseldorf


Dr. Maria Reich

maria.reich@uni-duesseldorf.de

Universitätsklinikum Düsseldorf


Prof. Dr. Mathias Heikenwälder

m.heikenwaelder@dkfz-heidelberg.de

DKFZ Heidelberg




QBiC contacts:

Dr. Gisela Gabernet

Bioinformatics project manager

gisela.gabernet@qbic.uni-tuebingen.de

1 Introduction

2 Loading the dataset

Loading all the individual samples.

Merging all samples in the dataset. The table represents the number of cells that are available for each sample.

Condition Cell_number
216-01 5300
216-02 4919
216-03 7252
216-04 6426
216-05 9975
216-09 7520
216-10 3498
216-11 9557

Adding sample conditions. The table represents the number of cells that are available for each condition.

Condition Cell_number
MDR2_KO 28953
TGR5_high 25494

3 Standard pre-processing workflow

Important QC params for eliminating bad quality cells (could be droplets without cells) are:

  • Number of unique genes detected in each cell
  • Total number of molecules detected within a cell

Calculating the percentage of genes mapping to mitochondrial genome for QC:

  • Low-quality / dying cells often exhibit extensive mitochondrial contamination
  • We calculate mitochondrial QC metrics with the PercentageFeatureSet function, which calculates the percentage of counts originating from a set of features
  • We use the set of all genes starting with mt- as a set of mitochondrial genes

Visualization of the QC metrics:

  • recommended to filter out the cells that have unique feature counts over 2500 or less than 200
  • recommended to filter out cells that have >10% mitochondrial counts

3.1 Filtering out low quality cells

Low quality cells need to be filtered out:

  • Filtering out the cells that have unique feature counts over 5,500 or less than 200
  • Filtering out cells that have >5% mitochondrial counts

After filtering low quality cells, the cell numbers per sample are the following:

Condition Cell_number
216-01 4276
216-02 4442
216-03 6359
216-04 5768
216-05 8795
216-09 6861
216-10 2876
216-11 8387

Violin plot after filtering

3.2 Data normalization

By default, we employ a global-scaling normalization method "LogNormalize" that normalizes the feature expression measurements for each cell by the total expression, multiplies this by a scale factor (10,000 by default), and log-transforms the result.

3.3 Detection of highly variable features

Detection of highly variable features. The displayed gene names represent the top 10 most variable genes.

## When using repel, set xnudge and ynudge to 0 for optimal results

3.4 Linear dimensionality reduction

Applies PCA to the highly variable features, after a standard scaling of the features to a mean of 0 and SD of 1.

## Centering and scaling data matrix
## PC_ 1 
## Positive:  Psap, Cst3, Ctsb, Ctsd, Ifi27l2a, Crip1, Zeb2, Lgmn, Fcer1g, Axl 
##     Hexa, Ctss, Cd74, Pltp, Laptm5, Klf2, Tyrobp, Csf1r, Rgs2, Gngt2 
##     C1qc, Lyz2, C1qa, Srgn, Ifi203, Aif1, C1qb, Cotl1, Mndal, Lst1 
## Negative:  Aldob, Fbp1, Chchd10, Ttc36, Stard10, Gsta3, Serpina1a, Bhmt, Ttr, Apoc4 
##     Cdo1, Rida, Apoc3, Urah, Gnmt, Serpina1b, Rbp4, Serpina1c, Hpd, Apoa2 
##     Ass1, Tdo2, Fabp1, Otc, Serpina1d, Uox, Phyh, Orm1, Car3, Angptl3 
## PC_ 2 
## Positive:  Ctss, C1qc, Tyrobp, C1qa, C1qb, Laptm5, Aif1, Wfdc17, Mpeg1, Cybb 
##     Lyz2, Ctsc, Csf1r, Fcer1g, Lst1, Cd300c2, Lcp1, Adgre1, Ccl6, Fyb 
##     Cd5l, Fcgr3, Ly86, Clec4f, Ptprc, Vsig4, Rgs1, Cd52, Cd68, Cfp 
## Negative:  Sparc, Igfbp7, Serpinh1, Bgn, Col3a1, Col1a2, Ccdc80, Col6a1, Col1a1, Dcn 
##     Col14a1, Col6a2, Cygb, Pcolce, Fstl1, Aebp1, Lum, Loxl1, Htra1, Adamts2 
##     Ltbp4, Col6a3, Dpt, Prelp, Col5a2, Lhfp, Fbln1, Serping1, Abi3bp, Gas6 
## PC_ 3 
## Positive:  Epcam, Cd24a, Sorbs2, Krt8, Sox9, Tspan8, Plet1, Krt19, Ehf, Sftpd 
##     Mfge8, Chka, Fxyd3, Wfdc2, Ddit4l, Krt23, Tinagl1, Tnfrsf12a, Ccn1, Rgs5 
##     Muc1, Plscr1, Atf3, Ankrd1, Hbegf, Krt7, Kcne3, Klf5, Tuft1, Clcf1 
## Negative:  Col1a2, Col3a1, Col1a1, Lum, Dcn, Dpt, Col6a1, Cygb, Mmp2, Col6a2 
##     Fbln1, Cfh, Mfap4, Islr, Adamts2, Colec12, Loxl1, Col14a1, Htra3, Ptgis 
##     Lrp1, Ms4a4d, Rarres2, Cxcl12, Serpinf1, Emp3, Abi3bp, Pdgfra, Gsn, Col5a1 
## PC_ 4 
## Positive:  Epcam, C3, Krt8, Sox9, Efemp1, Cd24a, Krt19, Plet1, Tspan8, Fn1 
##     Sod3, Tmem45a, Itih5, Fxyd3, Ehf, Sftpd, Tnfrsf12a, Wnt4, Wfdc2, Chka 
##     Atf3, Gas6, Krt23, Ddit4l, Sorbs2, Krt7, Rgs5, Gadd45b, Muc1, Ankrd1 
## Negative:  Ptprb, Adgrf5, Fabp4, Aqp1, Pecam1, Adgrl4, Kdr, Tek, Mmrn2, Cyyr1 
##     Clec14a, Flt1, F8, Jam2, Calcrl, Ehd3, Egfl7, Gpihbp1, Esam, Eng 
##     Ramp2, Plvap, Emcn, Myct1, Tie1, Pde2a, Gpr182, Clec4g, Cemip2, Vwf 
## PC_ 5 
## Positive:  Sparcl1, Des, Myl9, Lmod1, Gm13889, Itga8, Mylk, Atp1a2, Mustn1, Pi15 
##     Tagln, Gucy1b1, Mamdc2, Tbx2, Emilin1, Prrx1, Abcc9, Pde1a, Mfap4, Hand2os1 
##     Ebf1, Myh11, Cxcl12, Lum, Pdgfrb, Rgs7bp, Crispld2, Gem, Olfml3, Lbh 
## Negative:  Upk3b, Upk1b, Msln, Lrrn4, Cldn15, Rspo1, Nkain4, Igfbp6, Spock2, Muc16 
##     Bst1, Bnc1, Sulf1, Lrp2, Pkhd1l1, Slc16a1, Gpc3, Atp6v0a4, Igfbp5, Tmem151a 
##     Stk26, Myl7, Adgrd1, Chst4, Cybrd1, Clic3, Cyp2s1, Slc39a8, Prss12, Enpp6

## Saving 7 x 5 in image
## Saving 7 x 5 in image

3.5 Determine dimensionality of data set

Determining number of PCAs to consider for clustering from Elbow plot. It is recommended to go rather on the higher end of PCAs.

## Saving 7 x 5 in image
## Saving 7 x 5 in image
  • Number of chosen PCs: 15.

4 Cell clustering

As in PhenoGraph, we first construct a KNN graph based on the euclidean distance in PCA space, and refine the edge weights between any two cells based on the shared overlap in their local neighborhoods (Jaccard similarity).

This step is performed using the FindNeighbors function, and takes as input the previously defined dimensionality of the dataset.

To cluster the cells, we next apply modularity optimization techniques such as the Louvain algorithm (default) or SLM, to iteratively group cells together, with the goal of optimizing the standard modularity function.

## Computing nearest neighbor graph
## Computing SNN
  • Number of chosen PCs: 15.
  • Chosen resolution: 0.2.

4.1 UMAP: Non-linear dimensionality reduction

Non-linear dimensionality reduction method UMAP was applied to visualize the cell clusters.

## 22:09:01 UMAP embedding parameters a = 0.9922 b = 1.112
## 22:09:01 Read 47764 rows and found 15 numeric columns
## 22:09:01 Using Annoy for neighbor search, n_neighbors = 30
## 22:09:01 Building Annoy index with metric = cosine, n_trees = 50
## 0%   10   20   30   40   50   60   70   80   90   100%
## [----|----|----|----|----|----|----|----|----|----|
## **************************************************|
## 22:09:09 Writing NN index file to temp file /var/folders/2z/rqg0lktx6nq0_wqgs1yrx0lw0000gn/T//Rtmpcslb0s/file1487437202936
## 22:09:09 Searching Annoy index using 1 thread, search_k = 3000
## 22:09:28 Annoy recall = 100%
## 22:09:29 Commencing smooth kNN distance calibration using 1 thread
## 22:09:31 Initializing from normalized Laplacian + noise
## 22:09:44 Commencing optimization for 200 epochs, with 2008542 positive edges
## 22:10:11 Optimization finished

## Saving 7 x 5 in image
## Saving 7 x 5 in image

4.2 Find differentially expressed features

Finding markers (genes) that define clusters via differential gene expression expression. A table containing the top markers found for each of the clusters can be found under results/markers/.

## Calculating cluster 0
## Calculating cluster 1
## Calculating cluster 2
## Calculating cluster 3
## Calculating cluster 4
## Calculating cluster 5
## Calculating cluster 6
## Calculating cluster 7
## Calculating cluster 8
## Calculating cluster 9
## Calculating cluster 10
## Calculating cluster 11
## Calculating cluster 12
## Calculating cluster 13
## Calculating cluster 14
## Calculating cluster 15
## Calculating cluster 16
## Calculating cluster 17

Expression heatmap for the top 20 markers for each cluster. All plots for the markers can be found under results/markers.

4.3 Plotting provided markers

All marker plots in this section can be found under results/markers.

4.3.1 Markers for the cell types

Choose the cell type:

Cholangiocytes

Macrophage

Kupffer cells

Hepatic Stellate Cells (HSC)

Provided Hepatic stellate cells markers: Des, Reln. Other markers in this cluster: Dcn, Gsn, Cxcl12, Lum

Endothelial cells

Fibroblasts

Hepatocytes

## Warning in FetchData(object = object, vars = c(dims, "ident", features), : The
## following requested variables were not found: Serpina 1b

Liver projenitor cells

Plotting provided liver progenitor cells (LPC) markers.

B-lymphocytes

T-lymphocytes

5 Cluster cell type assignment

UMAP plot with cell type assignments according to gene expression profiles of the clusters. All UMAP plots are found under results/umap.

## Saving 7 x 5 in image
## Saving 7 x 5 in image

Number of cells identified for each of the cell types:

Condition Cell_number
Cholangiocytes 16060
Macrophage / Kupffer 10284
Hepatocytes 12306
HSC 2880
Endothelial 3426
T-lymphocytes 844
Endothelial vascular 651
Fibroblasts 590
B-lymphocytes 434
Lymphocytes 289

Heatmap with new cluster labels:

Samples

Grouping

Grouping split

6 Differential gene expression analysis

The differential gene expression analysis was performed for each of the cell types, by comparing the MDR2 KO to the WT mice population. Here the results tables are displayed for each cell population.

All tables for the Differential Gene Expression analysis can be found under results/DE_genes.

Cholangiocytes

p_val avg_logFC pct.1 pct.2 p_val_adj gene log_pval_adj
Pnkd 0 0.6354260 0.882 0.650 0 Pnkd Inf
Tmbim1 0 0.5556900 0.657 0.323 0 Tmbim1 Inf
Aamp 0 0.4378888 0.818 0.650 0 Aamp Inf
Ddx3y 0 0.3382718 0.157 0.000 0 Ddx3y Inf
Hspa8 0 -0.3110741 0.964 0.989 0 Hspa8 Inf
Xist 0 -0.4108656 0.649 0.985 0 Xist Inf
Serpina10 0 -0.8454248 0.195 0.553 0 Serpina10 Inf
Serpina6 0 -1.5215662 0.048 0.385 0 Serpina6 Inf
mt-Nd5 0 0.3964567 0.964 0.921 0 mt-Nd5 265.8403
Pigr 0 0.7072268 0.604 0.396 0 Pigr 236.5766
Krt8 0 -0.3640415 0.986 0.995 0 Krt8 220.9421
Fabp1 0 0.4103162 0.495 0.272 0 Fabp1 180.6191
Igfbp4 0 -0.3751494 0.368 0.581 0 Igfbp4 174.5309
Malat1 0 0.2795960 0.999 0.998 0 Malat1 154.8080
Kcne3 0 0.4217557 0.674 0.506 0 Kcne3 152.9720
Gm42418 0 0.3588910 0.979 0.954 0 Gm42418 152.1815
Dynll1 0 -0.2758041 0.698 0.835 0 Dynll1 142.9373
Sparc 0 -0.2862455 0.822 0.907 0 Sparc 139.2745
Actg1 0 -0.2897031 0.991 0.995 0 Actg1 136.4482
Txnip 0 0.3510106 0.833 0.736 0 Txnip 136.1171 
## When using repel, set xnudge and ynudge to 0 for optimal results

HSC

p_val avg_logFC pct.1 pct.2 p_val_adj gene log_pval_adj
Tmbim1 0 0.6672841 0.725 0.389 0 Tmbim1 105.52471
Pnkd 0 0.4730268 0.787 0.499 0 Pnkd 79.57374
Aamp 0 0.4249373 0.776 0.533 0 Aamp 67.99756
Ddx3y 0 0.2949511 0.114 0.000 0 Ddx3y 40.00707
B2m 0 0.4272555 0.941 0.899 0 B2m 32.55355
mt-Nd4l 0 0.3163706 0.915 0.812 0 mt-Nd4l 32.49410
H2-D1 0 0.5592761 0.912 0.846 0 H2-D1 30.46005
Ubc 0 -0.3088392 0.899 0.951 0 Ubc 30.02219
Spp1 0 -0.8331191 0.455 0.597 0 Spp1 26.33048
mt-Nd5 0 0.2755401 0.912 0.826 0 mt-Nd5 24.61970
Anxa1 0 0.3235805 0.768 0.614 0 Anxa1 21.38562
Rps2 0 -0.3303674 0.507 0.652 0 Rps2 18.55778
Efemp1 0 0.2982785 0.573 0.397 0 Efemp1 17.56830
Sparc 0 -0.2832436 0.993 0.988 0 Sparc 17.28275
mt-Nd3 0 -0.2531829 0.867 0.871 0 mt-Nd3 15.09504
Gstm1 0 0.2682395 0.936 0.883 0 Gstm1 13.92353
Penk 0 0.3607909 0.274 0.146 0 Penk 13.65636
Jund 0 0.3125016 0.957 0.933 0 Jund 13.31948
Ogn 0 0.3431669 0.677 0.551 0 Ogn 11.36462
H2-K1 0 0.3373740 0.898 0.863 0 H2-K1 10.85760 
## When using repel, set xnudge and ynudge to 0 for optimal results

Fibroblasts

p_val avg_logFC pct.1 pct.2 p_val_adj gene log_pval_adj
Mpp6 0e+00 -0.7901515 0.168 0.590 0.0000000 Mpp6 24.993789
Tmbim1 0e+00 0.5207181 0.704 0.369 0.0000000 Tmbim1 15.260526
B2m 0e+00 0.5200070 0.980 0.962 0.0000000 B2m 15.094202
Aamp 0e+00 0.4475090 0.808 0.570 0.0000000 Aamp 13.777739
Pnkd 0e+00 0.5055122 0.751 0.498 0.0000000 Pnkd 13.637931
Sparc 0e+00 -0.4148776 0.993 0.993 0.0000000 Sparc 9.885798
H2-D1 0e+00 0.5990800 0.953 0.959 0.0000000 H2-D1 7.615496
Ang 0e+00 0.3831581 0.488 0.218 0.0000001 Ang 7.202999
Ddx3y 0e+00 0.2747410 0.155 0.000 0.0000001 Ddx3y 7.087120
Xist 0e+00 -0.4704138 0.461 0.737 0.0000005 Xist 6.295809
Ubc 0e+00 -0.3773012 0.916 0.952 0.0000009 Ubc 6.029792
Rarres2 0e+00 0.3793868 0.990 1.000 0.0000011 Rarres2 5.978549
H2-K1 0e+00 0.3990797 0.963 0.983 0.0000047 H2-K1 5.327477
Rnase4 0e+00 0.2731783 0.970 0.973 0.0000171 Rnase4 4.767019
Rps2 0e+00 -0.3525257 0.552 0.751 0.0000400 Rps2 4.397639
Serpinh1 0e+00 -0.3024662 0.902 0.962 0.0001277 Serpinh1 3.893836
2010001K21Rik 0e+00 0.2782936 0.448 0.218 0.0001511 2010001K21Rik 3.820779
mt-Nd5 0e+00 0.3242000 0.936 0.870 0.0003312 mt-Nd5 3.479887
Spp1 0e+00 -0.5913436 0.414 0.577 0.0003561 Spp1 3.448455
Serpinb6b 1e-07 -0.4237194 0.758 0.843 0.0019910 Serpinb6b 2.700937 
## When using repel, set xnudge and ynudge to 0 for optimal results

Macrophage / Kupffer

p_val avg_logFC pct.1 pct.2 p_val_adj gene log_pval_adj
Aamp 0 0.6254687 0.744 0.477 0 Aamp Inf
Fau 0 -0.3061068 0.964 0.990 0 Fau 273.3556
Rps15a 0 -0.3144404 0.943 0.973 0 Rps15a 244.2381
Fabp1 0 1.6037711 0.536 0.237 0 Fabp1 241.4718
Rps11 0 -0.3361929 0.929 0.967 0 Rps11 231.9247
Rpl32 0 -0.3249804 0.924 0.969 0 Rpl32 230.4496
Rpl27a 0 -0.2827200 0.944 0.972 0 Rpl27a 214.2065
Rpl34 0 -0.2959929 0.931 0.971 0 Rpl34 213.6860
Rpl39 0 -0.2777258 0.953 0.977 0 Rpl39 204.5159
Rpl37 0 -0.2659561 0.953 0.975 0 Rpl37 171.0503
Cd83 0 -0.5731607 0.524 0.730 0 Cd83 168.2408
Slc11a1 0 0.5110704 0.748 0.677 0 Slc11a1 167.9841
Rpl18a 0 -0.2811977 0.921 0.969 0 Rpl18a 167.2244
Rps5 0 -0.2596402 0.924 0.969 0 Rps5 167.1390
Cd52 0 -0.3801953 0.839 0.938 0 Cd52 165.1401
Pnkd 0 0.4372257 0.544 0.322 0 Pnkd 157.7877
Dusp2 0 -0.5773107 0.277 0.515 0 Dusp2 148.1814
Rpl6 0 -0.2611276 0.924 0.964 0 Rpl6 144.8840
Rpl28 0 -0.2722368 0.901 0.963 0 Rpl28 142.5081
Rpl36 0 -0.2623005 0.890 0.951 0 Rpl36 138.5803 
## When using repel, set xnudge and ynudge to 0 for optimal results

Endothelial

p_val avg_logFC pct.1 pct.2 p_val_adj gene log_pval_adj
Tmbim1 0 0.7909171 0.740 0.363 0 Tmbim1 139.70856
Pnkd 0 0.5728983 0.719 0.407 0 Pnkd 90.86566
H2-D1 0 0.5341454 0.966 0.945 0 H2-D1 89.95554
H2-K1 0 0.4955551 0.965 0.932 0 H2-K1 85.10493
B2m 0 0.4529357 0.965 0.924 0 B2m 79.93364
Aamp 0 0.5422947 0.656 0.385 0 Aamp 73.10954
Rpl37a 0 -0.2750290 0.979 0.986 0 Rpl37a 60.55155
Rps29 0 -0.2561387 0.951 0.965 0 Rps29 39.57848
H2-Q6 0 0.3634977 0.416 0.205 0 H2-Q6 37.81407
H2-T23 0 0.3681642 0.828 0.687 0 H2-T23 34.21035
H2-Q7 0 0.3833302 0.499 0.302 0 H2-Q7 30.13458
Rps18 0 -0.3647375 0.456 0.653 0 Rps18 29.47065
Rps2 0 -0.3543401 0.444 0.623 0 Rps2 24.63197
Klf2 0 0.3584539 0.918 0.874 0 Klf2 24.29165
Ftl1 0 0.2592899 0.982 0.961 0 Ftl1 23.25587
mt-Nd4l 0 0.2743161 0.811 0.678 0 mt-Nd4l 19.27798
mt-Nd5 0 0.2634792 0.791 0.660 0 mt-Nd5 19.16932
Fxyd6 0 0.3093772 0.300 0.155 0 Fxyd6 18.58784
Sparc 0 -0.3100792 0.819 0.879 0 Sparc 18.54379
Xist 0 -0.2865103 0.729 0.855 0 Xist 18.05250 
## When using repel, set xnudge and ynudge to 0 for optimal results

Endothelial Vascular

p_val avg_logFC pct.1 pct.2 p_val_adj gene log_pval_adj
Aamp 0.0e+00 0.6124211 0.671 0.338 0.0000000 Aamp 15.5881326
Tmbim1 0.0e+00 0.6063314 0.621 0.257 0.0000000 Tmbim1 15.4778075
Pnkd 0.0e+00 0.5826073 0.706 0.421 0.0000000 Pnkd 13.5542901
Fabp1 0.0e+00 1.1575273 0.476 0.193 0.0000000 Fabp1 9.1618052
H2-D1 0.0e+00 0.3975316 0.965 0.971 0.0000003 H2-D1 6.4956070
Rpl37a 0.0e+00 -0.2507606 0.971 0.981 0.0000022 Rpl37a 5.6666301
Sparc 0.0e+00 -0.6113349 0.553 0.717 0.0000037 Sparc 5.4370717
Xist 0.0e+00 -0.3369773 0.803 0.932 0.0000050 Xist 5.3040000
Col4a2 0.0e+00 -0.4118840 0.482 0.666 0.0000174 Col4a2 4.7585431
Rps2 0.0e+00 -0.3977003 0.412 0.621 0.0002008 Rps2 3.6971962
Gstm1 0.0e+00 0.4673732 0.559 0.357 0.0006989 Gstm1 3.1556036
Fos 0.0e+00 0.4666698 0.865 0.733 0.0008460 Fos 3.0726062
Fxyd1 1.0e-07 0.3353177 0.524 0.309 0.0020677 Fxyd1 2.6845114
mt-Nd3 4.0e-07 -0.3843444 0.547 0.685 0.0135784 mt-Nd3 1.8671512
Nrp2 4.0e-07 -0.3962025 0.874 0.894 0.0137220 Nrp2 1.8625831
Stxbp6 1.0e-06 0.3224920 0.476 0.293 0.0331104 Stxbp6 1.4800360
H2-K1 1.4e-06 0.2763872 0.971 0.949 0.0449407 H2-K1 1.3473601
Stab2 1.5e-06 -0.4697030 0.291 0.453 0.0497519 Stab2 1.3031903
Gm38832 1.9e-06 -0.3548581 0.044 0.151 0.0607635 Gm38832 1.2163575
Nsg1 4.9e-06 0.3126703 0.544 0.360 0.1575257 Nsg1 0.8026487 
## When using repel, set xnudge and ynudge to 0 for optimal results

Hepatocytes

p_val avg_logFC pct.1 pct.2 p_val_adj gene log_pval_adj
Scd1 0 0.9245891 0.759 0.486 0 Scd1 Inf
Aox3 0 0.8987515 0.831 0.611 0 Aox3 Inf
Pnkd 0 0.5593918 0.836 0.663 0 Pnkd Inf
Aamp 0 0.5216661 0.801 0.578 0 Aamp Inf
Fabp1 0 0.4597307 0.998 0.985 0 Fabp1 Inf
Ftl1 0 0.4000707 1.000 0.997 0 Ftl1 Inf
Serpina12 0 1.0664946 0.626 0.354 0 Serpina12 288.4392
Mup21 0 1.3750889 0.355 0.073 0 Mup21 256.2281
Car3 0 0.5929682 0.987 0.964 0 Car3 222.9452
Ubc 0 -0.4238153 0.888 0.941 0 Ubc 193.3145
Elovl3 0 0.4935397 0.222 0.012 0 Elovl3 191.1047
Glud1 0 0.3261374 0.971 0.952 0 Glud1 177.1571
Insig2 0 -0.4151271 0.809 0.894 0 Insig2 173.4919
Akr1c19 0 -0.3618512 0.350 0.578 0 Akr1c19 167.1456
Nnmt 0 -0.4757936 0.756 0.838 0 Nnmt 166.8207
Hsd3b5 0 0.2951100 0.178 0.013 0 Hsd3b5 141.3752
Hpd 0 -0.2732340 0.983 0.967 0 Hpd 135.5089
Gm31583 0 0.4116757 0.356 0.147 0 Gm31583 135.1967
Rps2 0 0.4116527 0.761 0.668 0 Rps2 126.2926
1810008I18Rik 0 0.4620801 0.655 0.481 0 1810008I18Rik 125.1804 
## When using repel, set xnudge and ynudge to 0 for optimal results

Saving R Seurat object:

7 Methods

For the single-cell data analysis the R package Seurat v3.2.2 was employed. Graphs were produced in RStudio with R version 4.0.3 (2020-10-10) mainly using the R package ggplot2 v3.3.2. Final reports were produced using the R package rmarkdown v2.6, with knitr v1.30.